Obesity Surgery

https://doi.org/10.1007/s11695-019-03862-z

ORIGINAL CONTRIBUTIONS

Effectiveness of Ursodeoxycholic Acid in the Prevention

of Cholelithiasis After Sleeve Gastrectomy
Muriel Coupaye 1 & Daniela Calabrese 2 & Ouidad Sami 1 & Nathalie Siauve 3 & Séverine Ledoux 1

# Springer Science+Business Media, LLC, part of Springer Nature 2019

Abstract
Purpose The use of ursodeoxycholic acid (UDCA) to prevent gallstone formation after sleeve gastrectomy (SG) is still debated.
Furthermore, no study has assessed the effectiveness of UDCA on gallstone formation after the first postoperative year. Our aim
was to compare the incidence of cholelithiasis (CL) at 1 and 3 years after SG between patients treated or not treated with UDCA.
Materials and Methods From January 2008, a postoperative ultrasound monitoring was scheduled for all patients who underwent
SG in our institution. Patients with a preoperative intact gallbladder who performed at least one ultrasound at 1 year after SG were
included. We compared the incidence of CL between patients operated before October 2013 who did not receive UDCA and
those operated from October 2013 who received UDCA 500 mg once daily for 6 months postoperatively.
Results The incidence of CL at 1 year after SG was 28% in the 46 non-treated and 3.5% in the 143 treated patients (p < 0.001).
UDCA reduced the proportion of cholecystectomies from 11% to 1.4% (p = 0.012). Thus, the number of patients needed to treat
to avoid a cholecystectomy was about 10. Only 2 patients (1.4%) stopped UDCA for adverse effects. No gallstone appeared at 3
postoperative years in the 61 patients who performed an ultrasound at this time.
Conclusion UDCA 500 mg once daily for 6 months postoperatively is effective and well tolerated to prevent CL at midterm after
SG. We recommend UDCA treatment in all patients after SG with an intact preoperative gallbladder. However, large randomized
studies are needed to establish guidelines for prevention of gallstone formation after SG.

Keywords Sleeve gastrectomy . Gallstones . Cholelithiasis . Cholecystectomy . Ursodeoxycholic acid

Introduction the use of UDCA for 6 months after Roux-en-Y gastric bypass
(RYGB) to prevent gallstone formation [2, 3], but the preven-
Rapid weight loss in obese patients increases the risk of cho- tive strategy after sleeve gastrectomy (SG) is not established.
lelithiasis (CL). Ursodeoxycholic acid (UDCA) is a secondary Indeed, the effectiveness of UDCA to reduce the incidence of
bile acid that inhibits cholesterol secretion in bile, thereby CL after RYGB was demonstrated more than 20 years ago [4,
reducing cholesterol stone formation, especially in obese pa- 5], but few studies have explored the preventive effect of
tients [1]. American and French recommendations advocate UDCA after SG with heterogeneous results [6–8]. Abdallah
et al. recently showed that UDCA 600 mg per day for 6 months
reduced the incidence of gallstones from 5 to 0%, 1 year after
* Muriel Coupaye
SG in a retrospective study including more than 250 Egyptian
muriel.coupaye@aphp.fr
patients [8]. However, the very low incidence of postoperative
CL in this cohort makes this finding difficult to transpose in
1
Service des Explorations Fonctionnelles, Centre Intégré Nord other populations.
Francilien de prise en charge de l’Obésité (CINFO), Hôpital Louis
Mourier (AP-HP), Université Paris Diderot, Sorbonne Paris Cité, 178 We previously reported, in a prospective study, a similar
Rue des Renouillers, 92701 Colombes Cedex, France incidence of about one-third of CL after RYGB and SG in
2
Service de Chirurgie, Centre Intégré Nord Francilien de prise en patients without preventive treatment. In addition, more than
charge de l’Obésité (CINFO), Hôpital Louis Mourier (AP-HP), 10% of the patients with CL became symptomatic during
Université Paris Diderot, Sorbonne Paris Cité, Colombes, France follow-up after both procedures [9]. We thus decided to sys-
3
Service d’Imagerie Médicale, Hôpital Louis Mourier (AP-HP), tematically prescribe UDCA after RYGB and SG and showed
Université Paris Diderot, Sorbonne Paris Cité, Colombes, France that 500 mg UDCA once daily for 6 months reduced the
 OBES SURG

incidence of gallstone formation from 26% in 51 untreated Thus, all patients with a preoperative intact gallbladder
patients to 2.4% in 42 treated patients after SG followed for who performed at least one US at one postoperative year (±
1 year [7]. However, these results needed to be confirmed in a 3 months) or who were operated for cholecystectomy in the
larger cohort. Furthermore, to our knowledge, no study has first postoperative year were included. In these subjects, evo-
reported the incidence of gallstone formation in treated pa- lution of CL incidence from 6 months to 3 years (< 3.5 years)
tients more than 1 year after SG. was studied. Patients with early postoperative complications
The aims of our study were thus (1) to compare the inci- or who did not take UDCA for the entire period of 6 months
dence of gallstone formation using systematic abdominal ul- were excluded. The incidence of gallstones and sludge was
trasound (US) at 1 year in untreated and treated patients after compared between untreated subjects (from January 2008 to
SG, (2) to compare its evolution at 3 years after SG, and (3) to September 2013) and treated subjects (from October 2013)
assess the incidence of symptomatic patients requiring chole- after SG.
cystectomy in untreated and treated groups.

Statistical Analyses
Subjects and Methods
Data are expressed as means ± SDs for continuous variables
and as numbers or percentages for categorical variables.
Subjects and Surgical Procedure
Comparisons between groups were performed using
Student’s unpaired t tests or Mann–Whitney rank sum tests,
This observational study was based on our prospective data-
as appropriate, for continuous variables, and chi-squared tests
base including all obese patients referred for bariatric surgery
or Fisher’s exact tests, as appropriate, for categorical variables.
in our institution from 2004 [10]. This cohort was approved by
A p value < 0.05 was considered statistically significant.
our institution and local ethics committee; informed consent
Multiple linear regression analysis with presence or absence
was required and obtained from all patients before any inves-
of CL as a dependent variable was performed including the
tigations. Baseline characteristics were recorded for all pa-
patient’s characteristics exposed in Table 2 that have been
tients, as previously described [11]. Weight loss (WL) was
shown to be associated with CL in the literature. Statistical
expressed as absolute WL, % total weight loss (%TWL),
analyses were performed with SigmaPlot 12.5 (Systat
and % excess weight loss (%EWL). Laparoscopic SG was
Software, San Jose, CA).
performed using the same technique between 2008 and
2017, as previously described [11].

Investigations Results

A preoperative abdominal US, including the examination of Characteristics of the Patients

the gallbladder, was performed in all patients to rule out the
presence of gallstones or sludge. Cholecystectomy was per- Figure 1 depicts a flow diagram of the study population.
formed concomitantly only in case of symptomatic CL before Between January 2008 and August 2017, 305 SG were per-
SG. formed and followed in our institution. Eighty-two (27%) pa-
From January 2008, we decided to systematically detect tients who underwent SG were excluded because of preoper-
gallstone formation by US after bariatric surgery in order to ative gallbladder disease or previous cholecystectomy. Thus,
study the incidence of CL. An abdominal US, including the 223 patients had an intact gallbladder after SG and 32 patients
examination of the gallbladder, was scheduled at 6 months, were secondarily excluded for various reasons, including 22
1 year, and 3 years in all patients who underwent bariatric (9.9%) who did not perform any US at 1 year and 12 who did
surgery in our institution. In patients with a preoperative intact not take UDCA for 6 months (6 with severe postoperative
gallbladder, we started to systematically prescribe UDCA complications, 2 with adverse effects, 4 by omission).
500 mg once daily for 6 months after SG in October 2013. Finally, 189 patients were included in the analysis: 46 untreat-
The prescription of UDCA was driven by clinical consider- ed and 143 treated with UDCA (Fig. 1).
ations according to the high incidence of CL after both sur- The characteristics of the subjects prior to SG are shown in
geries [7]. We chose 500 mg once daily, just under the recom- Table 1. No difference was found between baseline character-
mended 600 mg per day after RYGB, because this corre- istics of the treated and untreated groups. In contrast, WL at
sponds to the most common conditioning available in 6 months after SG, when expressed as %TWL or %EWL, was
France and in order to increase the compliance, which has significantly higher in patients treated with UDCA than those
been reported as being low in this population [12–15]. without treatment (Table 1).
OBES SURG

Fig. 1 Flow diagram of the study

population. UDCA
ursodeoxycholic acid

Comparison of the Incidence of CL After SG significantly reduced by UDCA from 11 to 1.4% (p =

Between Treated and Untreated Groups 0.012). Thus, in non-treated subjects, near 40% of patients
with gallstones became symptomatic and required cholecys-
Among the 189 patients who performed an US at 1 year, 170 tectomy during follow-up. The 7 cholecystectomies (5 in non-
performed an US at 6 months and 61 at 3 years. The propor- treated subjects and 2 in treated subjects) were performed for
tion of patients followed at 3 years was not significantly dif- symptoms: 4 for biliary colic, 1 for acute cholecystitis, and 2
ferent in the treated group (63%; 42 of 67 patients operated for for symptomatic gallstone migration.
3 years or more) than in the untreated group (41%; 19/46
patients, p = 0.18). As shown in Fig. 2, postoperative inci- Predictive Factors for CL After SG in the Whole Cohort
dence of gallstones was significantly reduced from 28 to
3.5% by UDCA at 1 year after SG. About 60% of gallstones As shown in Table 2, the only preoperative variable sig-
present at 1 year were already present at 6 months. No gall- nificantly different between subjects without or with post-
stone appeared at 3 years in the 61 subjects who performed an operative CL was body mass index (BMI) > 50 kg/m2 that
US at 1 and 3 years (Fig. 2). was more frequent in subjects with CL. WL at 6 months
As shown in Fig. 2, the percentage of patients who was not significantly different between patients without or
underwent cholecystectomy during follow-up was with postoperative CL, but the number of subjects taking

Table 1 Subjects’ characteristics

Without UDCA With UDCA p value
(n = 46) (n = 143)

Men/women 6/40 24/119 0.71

Age (years) 43.2 ± 10.6 40.9 ± 11.2 0.18
Preoperative weight (kg) 122.2 ± 22.8 118.5 ± 21.2 0.39
Preoperative body mass index (kg/m2) 45.0 ± 8.4 43.3 ± 6.1 0.18
Mean weight loss at 6 months (kg) 25.6 ± 8.1 27.6 ± 7.9 0.17
Mean % total weight loss at 6 months (%) 20.8 ± 5.9 23.3 ± 5.5 0.02
Mean % excess weight loss at 6 months (%) 51.0 ± 19.1 57.9 ± 18.0 0.04
Baseline treatment for
Diabetes 7 (15) 15 (10) 0.55
Hypertension 13 (28) 35 (24) 0.75
Dyslipidemia 6 (13) 15 (10) 0.83
Sleep apnea syndrome 17 (37) 39 (27) 0.29

Values are means ± SDs or number of subjects (percentages). Data from subjects treated or not treated with UDCA
were compared using Student’s unpaired t test for continuous variables and the chi-squared test for categorical
variables
UDCA ursodeoxycholic acid
 OBES SURG

after SG. In addition to our previous study [7], two studies

reported incidence of CL in treated patients up to 1 year after
SG. The first was an American randomized study showing a
significant decrease in proportion of CL at 6 months (9% vs.
40%) in patients treated with UDCA 300 mg twice daily for
6 months, as compared with non-treated patients. However,
this finding was not significant at 1 year because of numerous
loss of follow-up [6]. The second study was a large Egyptian
retrospective study showing a reduction of gallstone forma-
tion from 5 to 0% at 1 year with UDCA 600 mg per day for
Fig. 2 Proportion of patients with cholelithiasis and cholecystectomy at
6 months, 1 year, and 3 years after sleeve gastrectomy according to
6 months after SG [8]. In the present study, UDCA 500 mg
treatment with ursodeoxycholic acid. NT non-treated subjects, T treated once daily for 6 months reduced gallstone formation from 28
subjects to 3.5% at 1 year after SG. Another important finding is that
no gallstone appeared between 1 and 3 years.
UDCA treatment was significantly lower in subjects with In non-treated patients, rapid weight loss has been reported
CL. In multivariate analysis, only absence of UDCA treat- to be the main risk factor for gallstone formation after SG [9,
ment (p < 0.001) was independently associated with gall- 15–17]. In the present study including a majority of treated
stone formation after SG. patients, only preoperative BMI > 50 kg/m2 was associated
with presence of CL, in contrast to the amount of weight loss
at 6 months. Finally, in accordance with Abdallah et al. [8], we
Discussion found that only absence of UDCA treatment was significantly
associated with presence of CL after SG in multivariate
We confirm in the present study that 500 mg UDCA once analysis.
daily for 6 months is very effective to prevent gallstone for- In addition, UDCA for 6 months reduced the number of
mation at 1 year but also at 3 years after SG. cholecystectomies for symptoms from 11 to 1.4% after SG in
In our knowledge, our study is the first to assess gallstone our study. Thus, the number of subjects needed to treat with
formation in treated and untreated subjects more than 1 year UDCA for 6 months after SG to avoid one cholecystectomy

Table 2 Factors associated with

cholelithiasis after sleeve n = 189 p value
gastrectomy
With CL Without CL
n = 18 n = 171

Age (years) 41.2 ± 12.4 41.5 ± 11.0 0.95

Gender (men/women) 4/14 26/145 0.66
Preoperative weight (kg) 129.5 ± 29.7 118.5 ± 20.4 0.16
Preoperative body mass index (kg/m2) 46.5 ± 11.2 43.5 ± 6.0 0.22
Preoperative body mass index > 50 kg/m2 6 (33) 20 (12) 0.03
Baseline treatment for
Diabetes 2 (11) 20 (12) 0.75
Hypertension 3 (17) 45 (26) 0.54
Dyslipidemia 1 (6) 20 (12) 0.69
Sleep apnea syndrome 7 (39) 48 (28) 0.49
Use of UDCA for 6 months 5 (28) 138 (81) < 0.001
Weight loss at 6 months (kg) 29.6 ± 8.8 26.8 ± 8.3 0.19
% total weight loss at 6 months (%) 22.8 ± 4.5 22.7 ± 7.1 0.93
% excess weight loss at 6 months (%) 54.7 ± 16.1 56.5 ± 22.7 0.81

Values are means ± SDs or number of subjects (percentages)

Data from subjects with CL and without CL were compared using the Mann–Whitney rank sum test for contin-
uous variables and the chi-squared test or Fisher’s exact test, as appropriate, for categorical variables
CL cholelithiasis, UDCA ursodeoxycholic acid
OBES SURG

was around 10. As in France, the cost of 6 months of UDCA Compliance with Ethical Standards
500 mg/day is about 150 euros per patient, and the cost of a
cholecystectomy is currently about 2500 euros per patient; this Conflict of Interest The authors declare that they have no conflicts of
interest.
treatment appears to be also cost-effective. Another benefit of
UDCA treatment after SG is probably a reduction in the risk of
Ethical Approval All procedures performed in studies involving human
acute pancreatitis since Hussan et al. recently reported a two- participants were in accordance with the ethical standards of the institu-
fold greater increase in the risk of acute pancreatitis within tional and national research committee and with the 1964 Helsinki dec-
6 months after SG compared with RYGB in a large laration and its later amendments or comparable ethical standards.
American database [18].
Informed Consent Informed consent was obtained from all individual
UDCA 500 mg once daily was very well tolerated in our participants included in the study.
study: only 2 patients (1.4%) stopped treatment for side effects
(allergy and nausea). The reasons for other patients who did
not take UDCA were surgical complications (n = 6) or lack of References
compliance (n = 4). Of note, UDCA treatment is fully reim-
bursed by the French public healthcare system, likely favoring 1. Mazzella G, Bazzoli F, Festi D, et al. Comparative evaluation of
the high compliance observed in our study. Furthermore, pa- chenodeoxycholic and ursodeoxycholic acids in obese patients.
tients were systematically informed on the expected risk of CL Effects on biliary lipid metabolism during weight maintenance
and weight reduction. Gastroenterology. 1991;101(2):490–6.
with and without treatment, which may also improve the
2. Mechanick JI, Youdim A, Jones DB, et al. Clinical practice guide-
compliance. lines for the perioperative nutritional, metabolic, and nonsurgical
However, there is still a debate concerning prevention and support of the bariatric surgery patient–2013 update: cosponsored
management of CL after SG. Indeed, in previous studies, the by American Association of Clinical Endocrinologists, the Obesity
Society, and American Society for Metabolic & Bariatric Surgery.
incidence of symptomatic CL in non-treated patients after SG
Obesity (Silver Spring). 2013;21(Suppl 1):S1–27.
was heterogeneous and usually lower than in the present 3. Haute autorité de santé [homepage on the Internet]. Obésité, prise
study: 7.5% in a large cohort including 361 patients [19], en charge chirurgicale chez l’adulte. Janvier 2009: www.has-sante.
around 6% in most studies [15, 20, 21], 2% in another study fr. Accessed January, 2, 2019.
4. Sugerman HJ, Brewer WH, Shiffman ML, et al. A multicenter,
[22], and near from 0% in two studies [8, 23]. As a result, most
placebo-controlled, randomized, double-blind, prospective trial of
authors do not advocate prophylactic cholecystectomy, use of prophylactic ursodiol for the prevention of gallstone formation fol-
UDCA treatment, or routine ultrasound surveillance after SG lowing gastric-bypass-induced rapid weight loss. Am J Surg.
[15, 19, 23, 24]. In our study, we found the same prevalence of 1995;169(1):91–6. discussion 96-7
about 10% of symptomatic gallstones after SG, as Altieri et al. 5. Magouliotis DE, Tasiopoulou VS, Svokos AA, et al.
Ursodeoxycholic acid in the prevention of gallstone formation after
[25] who followed 1650 patients after SG. Thus, we think as bariatric surgery: an updated systematic review and meta-analysis.
others [6, 8] that the high effectiveness and the good tolerance Obes Surg. 2017;27(11):3021–30.
of low-dose UDCA after SG justify the use of UDCA treat- 6. Adams LB, Chang C, Pope J, et al. Randomized, prospective com-
ment to prevent gallstone formation after SG. parison of ursodeoxycholic acid for the prevention of gallstones
after sleeve gastrectomy. Obes Surg. 2016;26(5):990–4.
Our study has two main limitations. First, it is not random- 7. Coupaye M, Calabrese D, Sami O, et al. Evaluation of incidence of
ized and the treated and untreated groups were not followed cholelithiasis after bariatric surgery in subjects treated or not treated
during simultaneous time periods. However, baseline charac- with ursodeoxycholic acid. Surg Obes Relat Dis. 2017;13(4):681–
teristics of both groups were similar and %TWL at 6 months 5.
8. Abdallah E, Emile SH, Elfeki H, et al. Role of ursodeoxycholic acid
was higher in the treated group, which enhances the effective- in the prevention of gallstone formation after laparoscopic sleeve
ness of UDCA treatment. Second, as in other studies, approx- gastrectomy. Surg Today. 2017;47(7):844–50.
imately 40% of patients were lost to follow-up at 3 years. 9. Coupaye M, Castel B, Sami O, et al. Comparison of the incidence
However, the follow-up was not significantly different be- of cholelithiasis after sleeve gastrectomy and Roux-en-Y gastric
bypass in obese patients: a prospective study. Surg Obes Relat
tween treated and untreated patients and no CL occurred after Dis. 2015;11(4):779–84.
the first postoperative year in subjects with US available at 10. Ledoux S, Calabrese D, Bogard C, et al. Long-term evolution of
3 years in our study. nutritional deficiencies after gastric bypass: an assessment accord-
In conclusion, UDCA 500 mg once daily for 6 months ing to compliance to medical care. Ann Surg. 2014;259(6):1104–
10.
postoperatively is very effective in preventing CL up to 11. Coupaye M, Rivière P, Breuil MC, et al. Comparison of nutritional
3 years after SG, is well tolerated, and appears to be cost- status during the first year after sleeve gastrectomy and Roux-en-Y
effective. We therefore recommend the use of UDCA in gastric bypass. Obes Surg. 2014;24(2):276–83.
all patients with an intact preoperative gallbladder to pre- 12. Scott DJ, Villegas L, Sims TL, et al. Intraoperative ultrasound and
prophylactic ursodiol for gallstone prevention following laparo-
vent gallstone formation after SG. However, large ran- scopic gastric bypass. Surg Endosc. 2003;17(11):1796–802.
domized studies are needed to establish guidelines for 13. Swartz DE, Felix EL. Elective cholecystectomy after Roux-en-Y
prevention of gallstone formation after SG. gastric bypass: why should asymptomatic gallstones be treated
 OBES SURG

differently in morbidly obese patients? Surg Obes Relat Dis. 20. Moon RC, Teixeira AF, Ducoin C, et al. Comparison of cholecys-
2005;1(6):555–60. tectomy cases after Roux-en-Y gastric bypass, sleeve gastrectomy,
14. Papasavas PK, Gagné DJ, Ceppa FA, et al. Routine gallbladder and gastric banding. Surg Obes Relat Dis. 2014;10(1):64–8.
screening not necessary in patients undergoing laparoscopic 21. Sioka E, Zacharoulis D, Zachari E, et al. Complicated gallstones
Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2006;2(1):41–7. after laparoscopic sleeve gastrectomy. J Obes. 2014;2014:468203.
15. Tsirline VB, Keilani ZM, El Djouzi S, et al. How frequently and 22. Mishra T, Lakshmi KK, Peddi KK. Prevalence of cholelithiasis and
when do patients undergo cholecystectomy after bariatric surgery? choledocholithiasis in morbidly obese South Indian patients and the
Surg Obes Relat Dis. 2014;10(2):313–21. further development of biliary calculus disease after sleeve gastrec-
16. Li VK, Pulido N, Fajnwaks P, et al. Predictors of gallstone forma- tomy, gastric bypass and mini gastric bypass. Obes Surg.
tion after bariatric surgery: a multivariate analysis of risk factors 2016;26(10):2411–7.
comparing gastric bypass, gastric banding, and sleeve gastrectomy. 23. Hasan MY, Lomanto D, Loh LL, et al. Gallstone disease after lap-
Surg Endosc. 2009;23(7):1640–4. Erratum in: Surg Endosc aroscopic sleeve gastrectomy in an Asian population-what propor-
2009;23:1645. Martinez-Duartez, Pedro tion of gallstones actually becomes symptomatic? Obes Surg.
17. Melmer A, Sturm W, Kuhnert B, et al. Incidence of gallstone for- 2017;27(9):2419–23.
mation and cholecystectomy 10 years after bariatric surgery. Obes 24. Li VK, Pulido N, Martinez-Suartez P, et al. Symptomatic gallstones
Surg. 2015;25(7):1171–6. after sleeve gastrectomy. Surg Endosc. 2009;23(11):2488–92.
25. Altieri MS, Yang J, Nie L, et al. Incidence of cholecystectomy after
18. Hussan H, Ugbarugba E, Porter K, et al. The type of bariatric sur-
bariatric surgery. Surg Obes Relat Dis. 2018;14(7):992–6.
gery impacts the risk of acute pancreatitis: a nationwide study. Clin
Transl Gastroenterol. 2018;9(9):179.
19. Dakour Aridi H, Sultanem S, Abtar H, et al. Management of gall-
bladder disease after sleeve gastrectomy in a selected Lebanese Publisher’s Note Springer Nature remains neutral with regard to jurisdic-
population. Surg Obes Relat Dis. 2016;12(7):1300–4. tional claims in published maps and institutional affiliations.